Tesla Silicon is Disrupting Silicon Disruption // + Notes on HPQ and Graphite
Welcome back everyone! I’m Jordan Giesige and this is The Limiting Factor. This is part seven of the Lithium Mine to Battery Line Series to break down and understand what was unveiled at Tesla Battery Day. Today we’re going to discuss how Tesla Silicon may disrupt the high loaded silicon anode market just as it’s about to get off the ground.
Before we begin, a special thanks to my Patreon supporters and YouTube members. This is the support that gives me the freedom to avoid chasing the algorithm and sponsors, and I hope will eventually allow me to do this full time. As always, the links for support are in the description. Back to Silicon disruption. In terms of battery chemistry, Silicon is one of the most promising ways to increase energy density and decrease the cost of lithium-ion battery cells. The more silicon you can safely get into the anode of the battery, the better. Before my final Battery Day Predictions video, I wrote multiple versions of the script. One of them included a low-cost and high energy density pathway that included Silicon.
However, the most promising silicon technology, silicon graphite composites, didn’t appear ready for primetime so I decided to leave them out of the script. Tesla’s Battery Day presentation referred to silicon graphite composites as silicon Structured in Graphite. Silicon Structured in Graphite and Silicon Graphite Composites are both a mouthful, so we’ll refer to them simply as composites. I started having doubts about composites
when I found out about SilLion, which Tesla allegedly acquired last year. SilLion specialised in the type of polymer coated silicon that Tesla revealed at battery day. But, I didn’t have any proof that SilLion was specifically working on the type of polymer coated silicon that could go into a Tesla vehicle, let alone whether they could get a product to market as quickly as a composite option. However, in the background, it was playing on my mind. A week after I made the SpaceX Moonshot video covering SilLion tech,
I raised the question with Shirley Meng. The reason why I didn’t release this footage with the original interview is because it would take an entire series on silicon, like the one I’m doing now, to provide background about what we were talking about and the significance. Have a listen. There’s a few technologies that I’ve been looking at. I don’t know if this is something you can comment on. It seems like Sila and Amprius…they both…Amprius actually has two technologies: The Silane Chemical Vapour Deposition, which is to me that’s too far off into the future. And they also have nano-composite and I think Sila also uses a nano-composite.
There’s a company that Tesla supposedly acquired called SilLion and they were working on Cyclised Polyacrylonitrile. If you were to hedge a bet on cyclised polyacrylonitrile vs like a nanocomposite like Sila has; Which direction do you think is the most promising? Wow, I think that’s a very difficult question! Yeah, so, I think that possibly the best way to look at this is that: What is going to increase the cost/performance ratio and safety at the same time? Based on my understanding of Tesla’s track record: Because they continuously increase the cost performance ratio, so basically… sorry, I should say performance per cost ratio… and you know at the same time keep up the safety record, right? So if I have to think about this…I do think any technology that uses a lot of fancy synthesis method, I don’t think it’s very difficult to lower that cost to performance. So, I think Tesla has the brilliant minds that making that kinds of decisions, so I’m confident that as long as they follow that guidance, I think that, you know, they make the right decision. Shirley’s general analysis turned out to be right on the money. First, Tesla focused on the cost to performance ratio of Tesla Silicon vs other Silicon products. Second, since battery day,
Elon has called out the importance of reliability and safety. As discussed in the last two videos, reliability and safety testing could mean that Tesla might need additional time to verify Tesla silicon, and therefore additional time to incorporate it into their batteries. Third, Tesla’s process for creating Tesla Silicon isn’t fancy and rather simple. In short, it involves dipping the silicon in a polymer solution and heat drying it.
With that said, reading between the lines, you can tell that we’re both struggling with the question. I can’t speak for Shirley but there were two reasons why I was finding hard to believe that Tesla would take the path of polymer coated Silicon particles. First, a graphite silicon composite particle is the approach that appeared to be the most likely for the entire industry. Several high profile start-ups are working on composites. Some of these companies are developing these composite particles to simply hedge their bet and cultivate a potential revenue source outside of their core product, this includes Talga and from what I can tell, Novonix.
Tesla Silicon isn’t a concern for me as an investor of both companies because I haven’t assumed that their growth will be built on silicon composites. In my view, the engine of growth for both these companies is graphite and other materials which I still expect Tesla and everyone else in the industry to continue using throughout the 2020s. Any revenue that Talga or Novonix make from composites will be icing on the cake. Any company that can’t develop a material like Tesla silicon or obtain it from Tesla will need their own high silicon anode material. They’ll be getting that high silicon material in the form of composites from companies like Talga and Novonix. However, other companies, which I won’t name, have staked their companies on silicon composites, and built their brand on them. The Silicon product
Tesla unveiled at battery day may have effectively knee capped many of these companies. Regardless, the primary point here is that the entire industry was and is gearing up for the composite approach. The second reason I had doubts about Tesla taking a polymer coated silicon approach before battery day was research. A broad strategy for increasing the Silicon content of batteries is detailed in this paper, which was published in May of this year. The title of the research paper was ‘The success story of graphite as a lithium-ion anode material – fundamentals, remaining challenges, and recent developments including silicon (oxide) composites.’
It was led by Jakob Asenbauer and overseen by Dominic Bresser. We’ll abbreviate the name of the paper to ‘The Success Story of Graphite Paper.’ I’ve outlined here the development path that the research paper laid out. Let’s look at what each of these three steps are and how Tesla has
flipped the board. The paper referred to them as steps, but I’ll refer to them as levels. As in level one to level three, with three being the most advanced. First level, Silicon Oxide Glass. This is what Tesla referred to in their Battery Day presentation as Silicon Structured in Silicon Oxide Glass. This is what manufacturers are currently using in their batteries to dope the graphite anode. The image on screen provides a better illustration of what the structure is like. The Silicon metal is encased in bulky silicon
oxide glass. However, as we discussed in the Cracking the Silicon Code video, silicon oxide glass appears to be limiting the silicon loading to 5-8% silicon. In order to take advantage of most of the benefits of Silicon, a loading of 20-30% is needed. Above 30% silicon, the voltage of the cell starts to drop, containing the volume expansion becomes more difficult, and there isn’t much graphite left in the anode to replace with higher energy density silicon. In short, there are diminishing returns and greater risks above 30% silicon in the anode.
The second level that the Success Story of Graphite Paper suggested was to use silicon-metal alloys. Silicon Metal Alloys appear to use similar engineering to silicon oxide glass, with two key differences. Instead of silicon surrounded by silicon oxide, they use silicon surrounded by silicon-metal alloys such as nickel and titanium. The nickel and titanium matrix has the additional benefit of being super elastic, similar to Tesla’s polymer coating. The silicon
metal alloy work showed a lot of promise and sounds like a fascinating solution to the lithium expansion problem. However, in my view, it would be better to skip this step. Like the silicon oxide glass, the metal matrix adds a lot of bulk to the Silicon and might reduce energy density. The Titanium-Nickel matrix would also likely be more expensive than Tesla’s polymer approach. The third level that the Success Story of Graphite Paper suggested was to use elemental silicon. Elemental silicon simply means pure silicon rather than an oxide or alloy. The paper listed several requirements for using pure silicon, which were as follows: First, limit the loading of silicon in the anode to 30-50% by weight, with the rest of the anode being filler material like graphite. The reason for this is straightforward. Silicon expands by 300% when charged and loading the anode past this can cause the anode material to disintegrate or peel away from the current collector. Second, limit the silicon particle size to
150 nanometres. This is because silicon pulverises itself above 150 nanometres. It’s also because electrode design and the elasticity of the binder come into play with larger micron sized particles. video Third, combine the 150 nanometer silicon particles with graphite particles to form larger spherical composite particles. Those larger spherical
conglomerates would be in the micron size range. This is the type of composite particle that Tesla showed in their presentation, which they referred to as silicon structured in graphite. My understanding is that combining the nano-particles into a larger micron composite particles provides two benefits. First, it creates a shared SEI layer that reduces first cycle losses. What are the SEI and first cycle losses? When a battery is first charged at the factory, the electrolyte reacts with the surface of the particles to form the Solid Electrolyte Interphase, or SEI, which is a protective layer that extends battery cycle life. The electrolyte contains lithium, which means that some of the lithium the battery uses to store energy is used to create the SEI layer. The second benefit is that these composite
particles are porous, which makes them more robust to the expansion and contraction of the Silicon particles. To summarise, as far as I can tell from the outside looking in at the battery industry and the research field, composite particles seemed like the most promising solution to both researchers and the battery industry for the following reasons: A maximum of 30-50% silicon loading makes the expansion problem manageable. 150 nanometre silicon eliminates silicon pulverisation. The composite particle lowers first cycle losses and the porous structure of the composite particles further buffers the silicon expansion. In my opinion, Tesla silicon is even more advanced than the level 3 solution that was called out in the research and I’d consider it a level 4 solution. Tesla Silicon uses a polymer coating that expands and contracts with the silicon as the battery charges and discharges. This solves multiple problems at once. First, it enables the use of silicon particles larger than 150nm. These larger silicon particles
still pulverise, but they do so within the protective and flexible polymer matrix. In fact, Tesla is probably using Silicon particles in the micron range, which is an order of magnitude larger than what would be used for composites. Micron sized particles have the additional benefit of being cheaper because they require less processing. Many people have been asking me to do a video on HPQ silicon or to give my opinion on HPQ.
This request appeared to be primarily pitched from an investor point of view. HPQ fits into the conversation here because it’s a good chance to talk about how much cheaper bulk silicon is than nano silicon of any grade. HPQ silicon claims to have developed a more environmentally friendly and cheaper way of producing silicon. One of the first things I noticed when looking into the company was that they were targeting the 150nm particles. If I’m right, then they’ve designed and marketed their product towards the particle sizes that will be used in silicon graphite composite anodes rather than Tesla Silicon.
However, HPQ updated their slide deck the day that I was finishing this script and they appear to have removed all mention of 150nm particles and instead state that they can create particles up to 5 microns in diameter. It looks like they may have recognised that they need to target a broader particle range. Battery Bulletin passed this chart to me which provides cost and grade information on silicon. For me, it tells the story of why HPQ would have targeted nano-silicon. The number of Ns indicates the number of nines. 2N for example is 99% purity and 5N is 99.999%. 99% purity, 2N silicon, which is represented by
MG for metallurgical grade, is $1.50 per kilogram. I’m not sure what the micronisation cost would be, but it should be significantly less than the cost for nano particles. Nano-silicon runs about $150 dollars per kilogram, regardless of purity. If the HPQ process produces nano-silicon in one step at 4N purity, which is what they’re claiming, and they can produce at costs comparable to non-nano silicon, they may get a lot of business from companies that produce composite particles. Personally, I’m rooting for Tesla’s approach,
which I think will use micron silicon at an order of magnitude lower cost than typical nano-silicon. HPQs high purity nano-silicon would be perfect for many other use cases, but if we take Tesla at their word and if I’m correct, they’re using the cheapest material out there which is micron sized metallurgical silicon. That is, Tesla targeted a universally available, low cost bulk material. Whereas HPQ might be better off targeting other markets with fatter margins
that require higher purity nano sized particles. Overall, HPQ appears to have great products that may be in demand for a number of specific use cases and they might help drive down the cost of the silicon that goes into batteries while offering a cleaner product. With that said, fitting with the theme of this video, Tesla’s approach is extremely disruptive, and we don’t know exactly what they’re using in their battery and their materials cost. This may be a big consideration if you’re an HPQ investor or considering it.
I should also take a moment here for the graphite bulls, which full disclosure, I am. Contrary to Silicon, battery grade graphite is scarce and uses proprietary production techniques. I’m guessing at most Tesla’s anodes will be 30% silicon by weight. That makes the other 70% graphite. Most battery producers will be using 90-95% graphite in their batteries for years to come. Additionally, the graphite industry has been expecting increases in silicon and built it into their growth models. The message here is that silicon won’t disrupt the battery materials market overnight, more graphite will be needed than silicon for the next decade, and unlike silicon there will likely be a shortage of graphite anode. As always, this isn’t investment advice.
This is my point of view and reflects my own investment strategy. If you don’t have your own well-researched opinions, the market’ll fleece you when you panic buy and panic sell. This is why you should always do your own research. Getting back on track, the second reason I view Tesla silicon as an advanced level 4 solution is that Tesla doesn’t need to fuse particles together into compact spheres. This should save time, energy, and money compared to composite particles. Along with that, the silicon and graphite can be added to the anode independently, increasing production flexibility at the point of production rather than up the supply chain.
Third, is speculation on my part. Tesla Silicon should have a smoother surface than the composite particles. A smoother surface should perform even better than composite particles when it comes to first cycle losses. The smoother the surface of a particle, the less surface area. In other words,
Tesla silicon should have less surface area due to both its larger size AND its shape, meaning less lithium is wasted in the first cycle. In summary, composite particles were set to disrupt the Silicon oxide glass currently used to dope Tesla’s graphite anode and the anodes of other large battery manufacturers. Tesla Silicon appears to be even simpler, more effective, and lower cost than composites. That is, Tesla may be disrupting composite particles just as the industry is lining up factories to produce battery cells with those composites. The composite particle is a brilliant idea.
However, Tesla may have one-upped it. The word first principles is getting tossed around a lot lately in relation to Tesla and some might feel the word is losing its meaning or impact. But, it is what it is. What other large company is brave enough to stray so far from the pack,
consistently identifies elegant solutions like polymer coated silicon, and grinds through the engineering required to make them work. At the beginning of this Lithium Mine to Battery Line Series I said that Tesla is taking big risks to pursue big rewards. Tesla Silicon is one of the biggest risks Tesla is taking and may provide some of the biggest rewards. While the rest of the industry will be using level 2 solutions like silicon metal alloys or level 3 solutions like composites, Tesla has created what I would consider a level 4 solution that’s better, faster, and cheaper. In the next video, we’ll wrap things up with the Tesla Silicon mini-series by covering myths about Tesla Silicon. This will be a tidy up before we press forward to wrap up the Lithium Mine to Battery Line Series by covering off the cathode, lithium clay, and structural batteries. If you enjoyed this video, please consider
supporting me on Patreon with the link at the end of the video or snag something off the merch shelf below. I am also active on Twitter and Reddit. You can find the details of those in the description and I look forward to hearing from you. A special thanks to Drew Cameron and Ryan for your generous support of the channel, all the other patrons listed in the credits, and my YouTube members. I appreciate all your support, and thanks for tuning in.
2021-01-03 00:12
